Business And IT Alignment For BI - Fact Or Fiction?

By Boris Evelson

My friend and highly respected colleague, Wayne Eckerson from TDWI, posted a great article called “Purple BI People”. In the article he described some of the best practices for business and IT alignment, and cross-functional skills sets needed for successful and effective BI professionals. Wayne, I loved the blue cow analogy, you know that I always think in metaphors, analogies, similies and associations. But, while I completely agree with Wayne in his near term assessment, best practices and recommendations, I would like to suggest another long term point of view.

Can business and IT ever align on BI? Can business ever be satisfied with IT for delivering successful and effective BI applications? Is there such a thing as BT (Business Technology, the phrase that Forrester coined and promotes) in BI?

I used to think we could deliver on that promise. Not so sure it’s that straightforward now. Just look at some of the hopelessly diametrically opposing business and IT priorities. I hear the following complaints from my clients day in and day out:

  • Business is all about revenue generation. While IT can support that, much more often cost cutting is IT's highest priority.
  • Business wants solutions now. Not tomorrow. IT needs to go through due diligence of testing and approving BI applications. Right now, on demand does not sit well with IT.
  • Business wants to react to constantly changing BI requirements. IT has to plan.
  • Business sometimes is willing to do something “quick and dirty” – even at the expense of potentially jeopardizing accuracy and adherence, compliance with standards. IT is all about compliance and sticking with standards.
  • Some business users, especially in the front office, will take speed and agility over a “single version of the truth”. I hear that loud and clear from some of my clients. But this is anathema for IT.

So can business and IT ever find the common ground, common language, lingua franca of BI? Wayne makes some great suggestions on how to do so. But – and this is only one poor analyst's opinion – perhaps the future lies in the clear separation of duties. Let’s fantasize for a second and imagine a bright future where we can reconcile the seemingly irreconcilable business and IT differences and invent a win-win situation as follows:

  • Let IT do what they do best: control, manage and plan. Let IT handle the so-called “lower layers” of the BI stack: data integration, data cleansing, MDM, metadata, data warehousing, etc. Just prepare the data for reporting and analysis.
  • Then, let business users write their own reports, ad-hoc queries, dashboards and perform whatever analysis they want. This is easier said than done, and this will require a huge cultural shift and many next-gen BI technologies. But some of these “next-gen” BI technologies needed to enable business users' self service do exist today and include data federation, SaaS, search-like UI, post-discovery, in memory and many others. They can indeed help to bring such a vision closer to reality.


What do you think? As always, more than interested in hearing the contrarian views.


Boris, In the New York Times


In the New York Times today (an unrelated article about whether going to college is still a good idea for many), was the following quote:

In one 2008 survey of more than 2,000 businesses in Washington State, employers said entry-level workers appeared to be most deficient in being able to “solve problems and make decisions,” “resolve conflict and negotiate,” “cooperate with others” and “listen actively.”

So this makes me wonder. We always perceive of people in organizations wanting to do well and working to further the organization's goals, so the antagonism between IT and the rest of the organization seems odd. Or is it? Read that paragraph again. Why would we assume that AFTER people come to work, they will magically train themselves to behave in a more productive manner? Maybe they never do.

I've long believed that the IT vs Business gap was artificial. After all, where would you find someone in an organization who defined their role as "the business?" It's distinction that was created by IT, an us vs. them mentality. Granted, people not in IT may envision the same partitioning of the world, but only from their own perch - IT vs us (marketing, etc.)

So maybe this is a sort of dark analysis of the problem, but the solution may be, given the rather poor skills described above, to stop relying on conversation to get people to work together, and wishful thinking, and maybe just kick people in the butt when they're out of line. Or teach them to be nice. On this latter point, I haven't a clue.

-Neil Raden

More comments

Some observations from the field

Hi Boris,

Returning from a customer visit in which many of these issues were apparent. The IT organization has begun to build federated metadata objects consisting of a variety of modern and woefully old, custom-built applications, with technology ranging from Oracle, Delphi, Powerbuilder, CA and so on. They are definitely seeking a single, consistent data model. On top of that, they have built a framework of dashboard, application and report content that integrates these systems and begins to create a consistent view of their performance. These representations roll up and down their org chart. IT is quite proud of their work, as they should be.

Their business users, Finance Heads and their direct reports, recognize that each individual has different thought patterns and preferences with regard to how they would like to receive their dashboard/performance content. Some like the gadgets and gauges, some don't trust them until they understand the underlying information (drill through is handy here). Some have better ideas about how to represent the information and want to jump in and change it…What they all agree on, however, is the need for some kind of starting point. Which brings me to my point regarding this particular topic. It's this, IT cannot stop at simply delivering the data or metadata model. They need to one step further and give the users a starting point from which to begin the "self serve" process. My favorite analogy here has always been that we all have made ice cream once in our lifetime… once. That's because it's too much of a hassle to repeat. Empty ad-hoc BI and query tools are the same way. The user will try it once, but figuring out which columns matter, and how to format the subsequent answer into a usable report is too much of a PITA for most business users. (If you're willing to repeat the process, you're a power user and not who I'm talking about.) So, we are seeing much better traction in not only creating the data, but also creating the initial, strawman report or dashboard content for the user, and then showing them how they can change and personalize it. Much like making ice cream sundaes at a sundae bar…Here every user wins, as each gets to choose their own toppings. These templates and half-built designs are a very useful aid in helping end users and IT huddle around their Business Intelligence and Reporting Tool. Then there's the ability to reuse and share what they have cooperatively built. That's cool.

Jeff Morris

Jeff, very thoughtful

Jeff, very thoughtful comments, I really appreciate it. Another way to look at this is front office vs. back office IT and business alignment. I do indeed see more structure and more alignment in the domains like finance and HR, where BI requirements are more structured and do not change on a dime. I also do not suggest in any way, shape or form, that, say, regulatory reporting should be left to the end users themselves. No, of course in any heavily regulated domain or where there’s lots of structure and ability to plan, traditional BI approaches (where business and IT do need to collaborate) still apply. However, sales and marketing, for example, are a completely different story. There we indeed see less structure (every sales deal may be structured differently) and a necessity to react instantaneously to, say, a new competitive threat. In those particular instances I do indeed see a clear separation of duties: just give business clean, trusted data and let them run their own analysis. This is actually already true today in many instances, especially in sales departments. I am aware of very few sales reporting and analytics environments build with traditional BI tools and approaches – they simply do not work.

Jeff said: "..but figuring

Jeff said: "..but figuring out which columns matter, and how to format the subsequent answer into a usable report is too much of a PITA for most business users. (If you're willing to repeat the process, you're a power user and not who I'm talking about.)"

Having been in IT involved in creating these template-oriented self-service reports you mention, I'm inclined to think the whole "too much of a PITA" is part of the problem.

Why is BI/data analytics so frequently shoved on to the basic ground-level biz user? I think there is a prevailing attitude out there that BI is "just reporting", and therefore best done by employees who are relatively low on the payscale.

Perhaps if companies hired people with backgrounds in mathematics, statistics, and the like, BI tools (even with all their shortcomings) would be used more effectively in those organizations. Yes, they'd have to pay more, but arguably, would extract more value from the reports, etc. those power users came up with.

Business and IT Alignment around BI

Hi: Boris:

This was a really interesting piece and close to our hearts at Ness. We learned some interesting things about Business and IT alignment around BI at some CIO Dinners we held over past few months. At the NYC at SF Dinners, BI partners from at least one company, a business person and an IT person, attended together. They were clearly in this BI thing together. A telling quote from a participant at the NYC Dinner was “I am optimistic about IT and business working together at my organization. We need this type of integration to ensure that the decisions we make around BI not only make sense from an IT perspective but are also strategic from a business perspective.” Other interesting learnings from our CIO Dinners are in our Topic Brief, “Recent Dialogues on Business Intelligence Reinforce Need for Real-Time Data and a Tight Connection between Business and IT Executives.”

Jennifer Scott
Ness Technologies

Jennifer, thanks for the

Jennifer, thanks for the comments. What you are describing is a goal of any organization that seeks effective and efficient BI solutions. It's BI 101. Unfortunately, it's not the reality in most cases. Our latest BI maturity survey findings only show modest, if that, levels of business and IT allignment in BI. And even in those cases, when I dig deeper, it's mostly just words.

it is necessary to have a best practice to deal with BI

I totally agree "Let IT do what they do best".

To let business users write their own reports, ad-hoc queries, dashboards, I doubt what kind of the business users they are, probably, Data Analyst, BA or the business analytic technology fans. Usually business users only focus on their business.

I like the win-win idea.


Flint Energy

BI Agility

Hi Boris,

I am really enjoying your various channels of thoughts on BI, it is great to get that helicopter view of it all, backed up by on the ground reality.

Thinking about BI in general there is a real philosophical question when the business asks IT for a specific report. Time and time again I get asked for a very complex report which allows multi dimensional analysis and predection as well as standard reporting.

When you actully break down their needs they are mixing their reports, and I think this is one of the fundamental reasons we get over complexity in BI reporting.

A lot of the time people are looking for a heads up that things are going ok. There is no point for someone working at groundlevel having a high level report showing the change of the team from one day to another. There is also no point asking for what-if analysis on all reporting because at the end of the day it is used by a few people infrequently.

More importantly when thinking about the what-if analysis, it is the limitations of the data which is just not understood when they make the requrest. In business when, for example, a marketing team want to analyse some data to see trends and patterns, a lot of time the underlying accuracy and sampling of the data is just not fine enought to give anything other than noise, or a normal bell shaped distribution. The idea that you can just bang in a few numbers and start predicting is I think the same impossible dream as being able to predict the stock exchange.

In the cases when there are large changes in the data from one period to another, so many times the underlying change is because of an external factor which cannot be controlled, and sometimes is only known and understood by the actual person who is dealing whith thoes clients. So when there is a large drop in revenue from a specific client, the client knows instantly that is because they down tools that month for an overhaul of the plant. This information was never, and will never be in the raw data.

I think that in the future, for businesses to retain the ability to stay agile they will need to give more power and control to the end users, and this will be done by providing generic signposting reporting to help them understand the general lay of the land, and then allow the users to augment their own high level data with their sector specific knowledge, which might include their own BI systems that will be much more ad-hoc and manula, so giving them more flexibility.


Thomas Salvini

Construqtive Limited